Chronic wound assessment and infection detection method.
Clustering
Edge detection
Image segmentation
Machine learning
Medical image processing
Surgical site classification
Wound assessment
Journal
BMC medical informatics and decision making
ISSN: 1472-6947
Titre abrégé: BMC Med Inform Decis Mak
Pays: England
ID NLM: 101088682
Informations de publication
Date de publication:
24 05 2019
24 05 2019
Historique:
received:
07
09
2017
accepted:
09
04
2019
entrez:
26
5
2019
pubmed:
28
5
2019
medline:
18
12
2019
Statut:
epublish
Résumé
Numerous patients suffer from chronic wounds and wound infections nowadays. Until now, the care for wounds after surgery still remain a tedious and challenging work for the medical personnel and patients. As a result, with the help of the hand-held mobile devices, there is high demand for the development of a series of algorithms and related methods for wound infection early detection and wound self monitoring. This research proposed an automated way to perform (1) wound image segmentation and (2) wound infection assessment after surgical operations. The first part describes an edge-based self-adaptive threshold detection image segmentation method to exclude nonwounded areas from the original images. The second part describes a wound infection assessment method based on machine learning approach. In this method, the extraction of feature points from the suture area and an optimal clustering method based on unimodal Rosin threshold algorithm that divides feature points into clusters are introduced. These clusters are then merged into several regions of interest (ROIs), each of which is regarded as a suture site. Notably, a support vector machine (SVM) can automatically interpret infections on these detected suture site. For (1) wound image segmentation, boundary-based evaluation were applied on 100 images with gold standard set up by three physicians. Overall, it achieves 76.44% true positive rate and 89.04% accuracy value. For (2) wound infection assessment, the results from a retrospective study using confirmed wound pictures from three physicians for the following four symptoms are presented: (1) Swelling, (2) Granulation, (3) Infection, and (4) Tissue Necrosis. Through cross-validation of 134 wound images, for anomaly detection, our classifiers achieved 87.31% accuracy value; for symptom assessment, our classifiers achieved 83.58% accuracy value. This augmentation mechanism has been demonstrated reliable enough to reduce the need for face-to-face diagnoses. To facilitate the use of this method and analytical framework, an automatic wound interpretation app and an accompanying website were developed. 201505164RIND , 201803108RSB .
Sections du résumé
BACKGROUND
Numerous patients suffer from chronic wounds and wound infections nowadays. Until now, the care for wounds after surgery still remain a tedious and challenging work for the medical personnel and patients. As a result, with the help of the hand-held mobile devices, there is high demand for the development of a series of algorithms and related methods for wound infection early detection and wound self monitoring.
METHODS
This research proposed an automated way to perform (1) wound image segmentation and (2) wound infection assessment after surgical operations. The first part describes an edge-based self-adaptive threshold detection image segmentation method to exclude nonwounded areas from the original images. The second part describes a wound infection assessment method based on machine learning approach. In this method, the extraction of feature points from the suture area and an optimal clustering method based on unimodal Rosin threshold algorithm that divides feature points into clusters are introduced. These clusters are then merged into several regions of interest (ROIs), each of which is regarded as a suture site. Notably, a support vector machine (SVM) can automatically interpret infections on these detected suture site.
RESULTS
For (1) wound image segmentation, boundary-based evaluation were applied on 100 images with gold standard set up by three physicians. Overall, it achieves 76.44% true positive rate and 89.04% accuracy value. For (2) wound infection assessment, the results from a retrospective study using confirmed wound pictures from three physicians for the following four symptoms are presented: (1) Swelling, (2) Granulation, (3) Infection, and (4) Tissue Necrosis. Through cross-validation of 134 wound images, for anomaly detection, our classifiers achieved 87.31% accuracy value; for symptom assessment, our classifiers achieved 83.58% accuracy value.
CONCLUSIONS
This augmentation mechanism has been demonstrated reliable enough to reduce the need for face-to-face diagnoses. To facilitate the use of this method and analytical framework, an automatic wound interpretation app and an accompanying website were developed.
TRIAL REGISTRATION
201505164RIND , 201803108RSB .
Identifiants
pubmed: 31126274
doi: 10.1186/s12911-019-0813-0
pii: 10.1186/s12911-019-0813-0
pmc: PMC6534841
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
99Références
IEEE Trans Med Imaging. 2010 Feb;29(2):410-27
pubmed: 19825516
Wound Repair Regen. 2009 Nov-Dec;17(6):763-71
pubmed: 19903300
IEEE Eng Med Biol Mag. 2007 Sep-Oct;26(5):18-22
pubmed: 17941318
IEEE Trans Neural Netw. 2002;13(2):415-25
pubmed: 18244442
Conf Proc IEEE Eng Med Biol Soc. 2004;2004:1389-92
pubmed: 17271952
Wounds. 2015 Oct;27(10):274-8
pubmed: 26479211
Conf Proc IEEE Eng Med Biol Soc. 2007;2007:6032-5
pubmed: 18003389
Int J Low Extrem Wounds. 2004 Sep;3(3):151-6
pubmed: 15866806
IEEE Trans Pattern Anal Mach Intell. 1986 Jun;8(6):679-98
pubmed: 21869365
IEEE Trans Med Imaging. 2011 Feb;30(2):315-26
pubmed: 20875969
Wounds. 2011 Sep;23(9):267-75
pubmed: 25879267
Biomed Res Int. 2014;2014:851582
pubmed: 25114925